The organ of Corti arises from a population of cells, referred to as prosensory cells, that are located within the inner ear. Based on existing data, prosensory cells are believed to be uniquely competent to develop as both hair cells and associated supporting cells. However, the factors that specify the prosensory domain remain unknown. In an ongoing project within the Section, we have examined the role of Sox2, a transcription factor that is known to play an important role in the development of the inner ear, along with other neural structures. Because deletion of Sox2 results in a lack of hair cells and supporting cells, it has been assumed that Sox2 acts to specify the prosensory population. However, using a gain-of-function approach, we have found that persistent expression of Sox2 actually acts to inhibit hair cell formation. Subsequent studies demonstrated that Sox2 actually works in concert with another transcription factor, Atoh1, to pattern the prosensory domain into hair cells and supporting cells. Sox2 actually inhibits the formation of hair cells, suggesting that its expression must be down regulated by Atoh1, before a hair cell can develop. These results suggest an intriguing reciprocal interaction between Sox2 and Atoh1 with each acting to antagonize the other, but with each also dependent on the other for expression. Therefore, it seems likely that the relative levels of expression of each factor within a single cell determine whether that cell will develop as a hair cell or a supporting cell.[unreadable] [unreadable] As discussed above, only a limited number of cells within the inner ear will develop as prosensory cells. However, an analysis of the evolution of the inner ear suggests that the number of cells within the ear that will develop as prosensory cells has actually decreased over evolutionary time. This observation suggests that inhibitory interactions may have been invoked to limit the size of prosensory populations. During the last year we have begun to examine signaling pathways that could play a role in inhibition of prosensory formation. We have identified one pathway; the Hedgehog signaling pathway, that we believe plays a role in limiting the extent of the prosensory domain. Using an in vitro system, we were able to demonstrate that increased activation of the hedgehog signaling pathway leads to a decrease in the size of the prosensory domain while inhibition of hedgehog signaling leads to increased prosensory cells. A similar result was observed in vivo in mice carrying a mutation in Gli3, a hedgehog target gene, which is also mutated in individuals with Pallister-Hall syndrome. Interestingly, Pallister-Hall patients were also found to have auditory defects, suggesting a role for hedgehog signaling both in mice and humans.[unreadable] [unreadable] The results of the first project described indicated that Sox2 plays an incompletely defined, but clearly crucial, role in formation of prosensory domains within the inner ear. In an effort to identify factors that act upstream of Sox2, we have examined the Notch signaling pathway as a possible regulator of Sox2. To address the role of Notch, we generated inner ear specific mutations in Rbp-j, a transcription factor that is required for all Notch signaling. We chose Rbp-j because multiple Notch genes are expressed in the ear, providing a possible mechanism for functional redundancy. The mammalian genome contains only a single Rbp-j gene, therefore deletion of this gene should completely inactive notch signaling. Results indicate that loss of Rbp-j results in a nearly complete loss of all hair cells and supporting cells within the inner ear. However, Sox2 expression is not directly affected by deletion of Rbp-j. This result suggests that while both Sox2 and Rbp-j are required for formation of hair cells and supporting cells, the two factors act through independent pathways. Based on these results it seems likely that multiple pathways may act in concert to regulate the formation of the sensory epithelium.[unreadable] [unreadable] In another series of studies, we examined the role of Sox2 in the formation of other aspects of the inner ear. In particular, we observed that Sox2 is also expressed in cells that will develop as the neurons that innervate and mechanosensory hair cells. These neurons coalesce to form the VIII cranial nerve, the statoacoustic nerve. The presence of these neurons is crucial for normal auditory function as well as for the success of cochlear implants. To determine whether Sox2 is required for formation of statoacoustic nerves, we analyzed inner ears from Sox2 mutant mice and observed a complete lack of neuronal innervation prior to the birth of these animals. To confirm the role of Sox2, we used gene transfer to force expression of Sox2 in cells within the inner ear that would not normally express this gene. As a result of forced expression of Sox2, these cells begin to express molecular markers of neuronal cells and begin to develop a morphology that is consistent with neurons. These results strongly suggest that Sox2 plays a role in the formation of neurons of the inner ear. [unreadable] [unreadable] Based on the previous results, we decided to examine whether other transcription factors that are expressed during the early formation of inner ear neurons could be used to induce cells within the inner ear to develop as neurons. Two transcription factors, Neurogenin1 and NeuroD1, were expressed in non-neuronal cells within the inner ear using gene transfer. Both of these factors were also able to induce the expression of neuronal markers and a neuronal phenotype. Moreover, co-expression of both Neurogenin1 and NeuroD1 resulted in the induction of neurons with a more mature phenotype. These result suggest that it may be possible to induce the formation of inner ear neurons using gene transfer. Such an approach could be used to develop therapies for the induction of new inner ear neurons in individuals in which these cells have been lost as a result of genetic mutation or trauma.